liveBook · Manning

Dialogue

Also called conversation

Dialogue is an ordered exchange of related messages across successive turns between participants or an interactive system and its users.

How 12 Manning books explain it

AI Agents and Applications

A chat message history is an ordered collection of user and language-model messages representing a conversation.

How it is used
Maintains context across chatbot and tool-using assistant turns so follow-up questions can refer to earlier exchanges.
How deep it goes
Works through message-based RAG prompts, history updates, and checkpoint-backed state restoration, including inspection of saved snapshots.
Examples
Uses a Poseidonia follow-up and a weather assistant’s “same town” question to demonstrate reference resolution.
What you can do after
Apply it via LangChain message history and LangGraph checkpoints to sustain multi-turn conversations.

Read how AI Agents and Applications explains dialogue →

Effective Conversational AI

Chat history is the recorded sequence of messages exchanged during a user's conversational interactions.

How it is used
Uses recorded user–bot exchanges to diagnose frustrating interactions, measure understanding, and support handoff summaries.
How deep it goes
Details log sampling, exchange-level organization, and intent annotation; gives summary prerequisites without a transcript-system implementation.
Examples
Illustrates a repeated “yes” response, first-utterance sampling, and comparisons between annotated intent volumes and training examples.
What you can do after
Evaluate/understand chatbot failures and intent accuracy using sampled, human-annotated dialogues while recognizing when later turns matter.

Read how Effective Conversational AI explains dialogue →

Build AI-Enhanced Web Apps

An ordered record of messages exchanged during a conversation with a user or conversational system.

How it is used
Maintains chat context for follow-up questions and lets users review or retrieve conversations.
How deep it goes
Explains client-server synchronization, session-based retrieval, and memory integration, while leaving storage integration to the developer.
Examples
Uses a scrollable chat interface and an interview assistant with Redis-backed session histories.
What you can do after
Apply it via LangChain.js by loading session messages into conversation memory and invoking a chain with the current query.

Read how Build AI-Enhanced Web Apps explains dialogue →

Designing AI Systems

A conversation is a structured sequence of messages exchanged between a user, an assistant, and the system instructions governing the interaction.

How it is used
Provides conversation continuity by retrieving prior messages for follow-up requests.
How deep it goes
Defines session and message contracts and implements a session-backed workflow, including a sliding-window trade-off.
Examples
Uses a patient intake follow-up, a retail return-policy question, and an assistant's restaurant recommendations.
What you can do after
Build a session-backed workflow that retrieves history, sends a conversation to a model, and stores each exchange.

Read how Designing AI Systems explains dialogue →

Data Analysis with LLMs

A message history is an ordered collection of conversational messages representing an interaction with a chat model.

How it is used
Supplies current and prior messages as context for chat replies and records desired interactions for fine-tuning.
How deep it goes
Defines message lists, dictionaries, roles, and content, then shows a completion call and labeled training histories without examining model internals.
Examples
Uses a story request, ChatGPT follow-ups, and a movie review paired with the desired `neg` label.
What you can do after
Apply it via the OpenAI Python library to submit message histories, or build labeled histories for fine-tuning.

Read how Data Analysis with LLMs explains dialogue →

Starting Data Analytics with Generative AI and Python

An interaction pattern in which multiple successive user and assistant messages form a single contextual conversation.

How it is used
Carries earlier prompts and responses into API requests so follow-up questions have exchange context.
How deep it goes
Defines message roles and history, and notes that requests must supply context because past requests are not remembered.
Examples
Uses a Python API exchange about explaining resonance to children, followed by a question about where resonance is used.
What you can do after
Apply dialogue by sending a message array containing prior user and assistant turns with each API request.

Read how Starting Data Analytics with Generative AI and Python explains dialogue →

LLMs in Production

Chat history is the ordered record of messages exchanged during a conversational interaction.

How it is used
Supplies prior user prompts and assistant responses as context for new prompts in an interactive LLM application.
How deep it goes
Demonstrates storing and displaying turns in Streamlit, notes token limits, and points to summarization for compression.
Examples
Shows a Streamlit chatbot storing role-tagged messages in session state and displaying earlier turns.
What you can do after
Apply it via Streamlit by storing chat turns in session state and using the history to form a new prompt.

Read how LLMs in Production explains dialogue →

Claude at Work

A conversation is a cumulative interaction session consisting of messages exchanged between a user and an AI assistant.

How it is used
Coordinates task requests, session context, outputs, and iterative evaluation with Claude.
How deep it goes
Explains cumulative token use and practical session management, but gives no detailed context-compression procedure.
Examples
Illustrates context growth through turns and tool-driven research, and recommends prompts structured around goals, context, constraints, and format.
What you can do after
Operate Claude sessions by structuring openings, revisiting or summarizing chats, and restarting confused threads.

Read how Claude at Work explains dialogue →

Build an AI Agent (From Scratch)

Conversation history is the accumulated sequence of prior messages retained for continuing an interaction with a language model.

How it is used
Maintains coherent interactions by resending prior messages or retrieving a session across agent runs.
How deep it goes
Implements message accumulation and session persistence, while identifying token growth as a cost and deferring efficient context management.
Examples
Demonstrates retaining a name across API calls and agent runs using a message list or session ID.
What you can do after
Apply continuity by resending prior messages or reusing a session ID across agent runs.

Read how Build an AI Agent (From Scratch) explains dialogue →

Grokking AI Applications

A record of earlier messages exchanged between a user and an agent or model.

How it is used
Preserves context between interactions by sending earlier messages with later prompts.
How deep it goes
Explains stateless calls and demonstrates a Python list that reinjects the conversation history at each turn.
Examples
Uses an Andrea name-recall exchange and a chatbot that appends prior turns to its prompt.
What you can do after
Apply it via a Python history list or Langflow’s Message History component to include earlier turns in later prompts.

Read how Grokking AI Applications explains dialogue →

Building Reliable AI Systems

A record of the messages exchanged during a conversation.

How it is used
Retains session context so an agent can reuse user details instead of asking for them again.
How deep it goes
Describes lightweight session storage and gives a simple list-based implementation.
Examples
Uses support-chat details, troubleshooting context, and an order ID to illustrate reuse during a session.
What you can do after
Build it from scratch by appending messages to a list and returning the stored history.

Read how Building Reliable AI Systems explains dialogue →

Build Applications with Local AI Models on a Mac

The example chatbot's ordered exchange of messages between a user and an AI assistant.

How it is used
Demonstrates how a chatbot uses conversation history to respond across multiple turns.
How deep it goes
Explains that the script sends full history and runs in a loop, without detailing history storage or message construction.
Examples
Uses a chatbot exchange where the assistant recalls a name and recommends a beginner programming book.
What you can do after
Operate a looping chatbot and end the conversation by typing quit, exit, or bye.

Read how Build Applications with Local AI Models on a Mac explains dialogue →

Also covered in

What to know first

Related concepts

Questions about dialogue

What is dialogue?

Dialogue is an ordered exchange of related messages across successive turns between participants or an interactive system and its users.

Which Manning books explain dialogue?

AI Agents and Applications, Effective Conversational AI, Build AI-Enhanced Web Apps, Designing AI Systems, Data Analysis with LLMs, Starting Data Analytics with Generative AI and Python, LLMs in Production, Claude at Work and 4 more.

What should I know before learning dialogue?

Manning books that teach dialogue usually introduce language model, conversational ai, context and prompt first.

What are the parts of dialogue?

Dialogue includes message and prompt.

What is dialogue also called?

Dialogue is also called conversation.

These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.